[Pdx-pm] Lisp class

Randal L. Schwartz merlyn at stonehenge.com
Sun May 1 08:33:00 PDT 2005


>>>>> "Michael" == Michael G Schwern <schwern at pobox.com> writes:

Michael> That said, I think what to pull out of Paul Graham's Lisp
Michael> assertion is not so much about Lisp persay but about
Michael> functional programming.

Lisp also provides a tiny syntax, and a self-referential syntax.  Lisp
programs can parse themselves, and generate themselves, fairly easily.
This means that meta-programming is the norm, not the exception.

The "scripting langauges" (languages that provide "eval" features) can
mimic some of this power as well.  I think that's why
Perl/Python/Ruby/et. al.  are re-capturing some of the programming
interest... because it's so easy to give the language the job of the
boring repetitive stuff that was formerly done by fancy ugly IDEs in
the 90s for languages like Java and C++.

Class::DBI does a lot of subroutine writing for me.  I can spell
things out once, and CDBI does the right thing to write dozens of
subroutines for me.  I can even add new recipes for creating
subroutines.  This is leverage.  But it's nothing new... Lisp was doing
this 40 years ago.

-- 
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<merlyn at stonehenge.com> <URL:http://www.stonehenge.com/merlyn/>
Perl/Unix/security consulting, Technical writing, Comedy, etc. etc.
See PerlTraining.Stonehenge.com for onsite and open-enrollment Perl training!


More information about the Pdx-pm-list mailing list